storium-backend
RL4LMs
| storium-backend | RL4LMs | |
|---|---|---|
| 4 | 5 | |
| 8 | 2,094 | |
| - | 2.2% | |
| 0.0 | 0.0 | |
| about 2 years ago | 2 months ago | |
| Python | Python | |
| BSD 3-clause "New" or "Revised" License | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
storium-backend
-
[R] Wordcraft: a Human-AI Collaborative Editor for Story Writing
I’m excited to see where research like this goes next. Though I’m biased considering my research on Storium.
-
[D] Very long sequence data (books) understanding?
I released a dataset of stories that are 19K tokens on average, but the longest are over a million. Our human evaluations show that relevance is the biggest factor in whether authors decide to use model generated text in their story, making this a good platform for assessing long document understanding and generation.
-
[P] Question about generating stories
More recent work tries to learn all of this purely from text. My dataset collected from Storium includes a narrator and annotations, e.g. challenges, goals, etc that can help learn these traits directly from the dataset.
-
[D] Deploying ML models - batching
If you’re willing to roll your own, you can see an example from my latest research project that makes use of asyncio.
RL4LMs
-
How To Setup a Model With Guardrails?
I think of guardrails as another dimension of human preferences: whether you are training a model to answer questions more gooder or avoid saying horrifying stuff, you are teaching the model a preference. So I thinks it's a straightforward RLHF problem but from a different perspective.
-
OpenDILab Awesome Paper Collection: RL with Human Feedback （2）
Found relevant code at https://github.com/allenai/rl4lms + all code implementations here
-
Best option for creating a custom GPT AI
If you skip down ther s several open source options. https://github.com/allenai/RL4LMs is an example.
- Will we ever see an open source alternative to ChatGPT?
- An Open-Source Version of ChatGPT is Coming [News]
What are some alternatives?
server - The Triton Inference Server provides an optimized cloud and edge inferencing solution.
trlx - A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
Cornucopia-LLaMA-Fin-Chinese - 聚宝盆(Cornucopia): 中文金融系列开源可商用大模型，并提供一套高效轻量化的垂直领域LLM训练框架(Pretraining、SFT、RLHF、Quantize等)
Dromedary - Dromedary: towards helpful, ethical and reliable LLMs.
Spectrum - Spectrum is an AI that uses machine learning to generate Rap song lyrics
PaLM-rlhf-pytorch - Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM
commit-autosuggestions - A tool that AI automatically recommends commit messages.
valhalla-nmt - Code repository for CVPR 2022 paper "VALHALLA: Visual Hallucination for Machine Translation"
GPT2-Chinese - Chinese version of GPT2 training code, using BERT tokenizer.
Open-Assistant - OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.
awesome-transformer-nlp - A curated list of NLP resources focused on Transformer networks, attention mechanism, GPT, BERT, ChatGPT, LLMs, and transfer learning.
NeMo-Guardrails - NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.